arXiv subjects

John Petkau

Publications and source records attributed to John Petkau.

4 recordsLinked to original sources

arXiv2026

Non-parametric assessment of the calibration of individualized treatment effects

An important aspect of the performance of algorithms that predict individualized treatment effects (ITE) is moderate calibration, i.e., the average treatment effect among individuals with predicted treatment effect of z being equal to z. The assessment of moderate calibration is challenging on two fronts: counterfactual responses are unobserved, and quantifying the conditional response function for models that generate continuous predicted values requires regularization. Perhaps because of these challenges, there is currently no inferential method for the null hypothesis that an ITE model is moderately calibrated in a population. In this work, we propose non-parametric methods for the assessment of moderate calibration of ITE models for binary outcomes using data from a randomized trial. These methods simultaneously resolve both challenges, resulting in novel graphical, numerical, and inferential methods for the assessment of moderate calibration. The key idea is to formulate a stochastic process for the cumulative prediction errors that obeys a functional central limit theorem, enabling the use of the properties of Brownian motion for asymptotic inference. We propose two approaches to construct this process from a sample: a conditional approach that relies on predicted risks (often an output of ITE models), and a marginal approach based on replacing the cumulative conditional moments with their marginal counterparts. Numerical simulations confirm the desirable properties of both approaches and their ability to detect miscalibration of different forms. We use a case study to provide suggestions on graphical presentation and the interpretation of results. Moderate calibration of predicted ITEs can be assessed without requiring regularization techniques or making assumptions about the functional form of treatment response. The accompanying cumulcalib R package implements this method.

stat.ME↗

arXiv2025

Identification of distributions for risks based on the first moment and c-statistic

We show that for any family of distributions with support on [0,1] with strictly monotonic cumulative distribution function that has no jumps and is quantile-identifiable (i.e., any two distinct quantiles identify the distribution), knowing the first moment and c-statistic is enough to identify the distribution. The derivations motivate numerical algorithms for mapping a given pair of expected value and c-statistic to the parameters of specified two-parameter distributions for probabilities. We implemented these algorithms in R and in a simulation study evaluated their numerical accuracy for common families of distributions for risks (beta, logit-normal, and probit-normal). An area of application for these developments is in risk prediction modeling (e.g., sample size calculations and Value of Information analysis), where one might need to estimate the parameters of the distribution of predicted risks from the reported summary statistics.

stat.ME↗

arXiv2024

Non-parametric inference on calibration of predicted risks

Moderate calibration, the expected event probability among observations with predicted probability z being equal to z, is a desired property of risk prediction models. Current graphical and numerical techniques for evaluating moderate calibration of risk prediction models are mostly based on smoothing or grouping the data. As well, there is no widely accepted inferential method for the null hypothesis that a model is moderately calibrated. In this work, we discuss recently-developed, and propose novel, methods for the assessment of moderate calibration for binary responses. The methods are based on the limiting distributions of functions of standardized partial sums of prediction errors converging to the corresponding laws of Brownian motion. The novel method relies on well-known properties of the Brownian bridge which enables joint inference on mean and moderate calibration, leading to a unified "bridge" test for detecting miscalibration. Simulation studies indicate that the bridge test is more powerful, often substantially, than the alternative test. As a case study we consider a prediction model for short-term mortality after a heart attack, where we provide suggestions on graphical presentation and the interpretation of results. Moderate calibration can be assessed without requiring arbitrary grouping of data or using methods that require tuning of parameters. An accompanying R package implements this method (see https://github.com/resplab/cumulcalib/).

stat.ME↗

arXiv2021

Model-based ROC (mROC) curve: examining the effect of case-mix and model calibration on the ROC plot

The performance of risk prediction models is often characterized in terms of discrimination and calibration. The Receiver Operating Characteristic (ROC) curve is widely used for evaluating model discrimination. When evaluating the performance of a risk prediction model in a new sample, the shape of the ROC curve is affected by both case-mix and the postulated model. Further, compared to discrimination, evaluating calibration has not received the same level of attention. Commonly used methods for model calibration involve subjective specification of smoothing or grouping. Leveraging the familiar ROC framework, we introduce the model-based ROC (mROC) curve to assess the calibration of a pre-specified model in a new sample. mROC curve is the ROC curve that should be observed if a pre-specified model is calibrated in the sample. We show the empirical ROC and mROC curves for a sample converge asymptotically if the model is calibrated in that sample. As a consequence, the mROC curve can be used to assess visually the effect of case-mix and model mis-calibration. Further, we propose a novel statistical test for calibration that does not require any smoothing or grouping. Simulations support the adequacy of the test. A case study puts these developments in a practical context. We conclude that mROC can easily be constructed and used to evaluate the effect of case-mix and model calibration on the ROC plot, thus adding to the utility of ROC curve analysis in the evaluation of risk prediction models. R code for the proposed methodology is provided (https://github.com/msadatsafavi/mROC/).

stat.ME↗